Characterization of the proteome changes in an in vitro hyperosmotic dry eye model employing human corneal epithelial cells

Abstract

Aims/Purpose: The detrimental effects of tear hyperosmolarity on human corneal epithelial (HCE) cells is a primary cause of dry eye syndrome (DES). However, the molecular mechanisms underlying inflammatory processes and cellular dysfunction in HCE are largely unknown. This study characterized the proteome changes in an optimized hyperosmolarity‐induced DES model in vitro.Methods: An in vitro hyperosmolar (400–550 mOsm) DES model was established in a simian virus 40 (SV40)‐immortalized HCE cell line. Cells incubated in medium with physiological osmolarity (312 mOsm) were used as control. Cell viability, cell death and intracellular reactive oxygen species (ROS) levels were analysed with MTS, Annexin‐V‐FLUOS, crystal violet and DCFDA assays. Mass spectrometry‐based proteomics analyses characterized the proteome and elucidated the underlying molecular changes.Results: The HCE cell viability was significantly decreased in a concentration‐dependent manner with increasing hyperosmolarity compared to control at 48 hr (450 mOsm, −47%, p = 2.7 × 10−8; 500 mOsm, −71%, p = 1.1 × 10−4; 550 mOsm, −84%, p = 2.6 × 10−5). Notably, cell adherence was significantly decreased (p = 5.8 × 10−9), while levels of ROS (p = 3.6 × 10−2) and apoptosis (p = 3.6 × 10−2) were significantly increased at 450 mOsm compared to control. Proteomics analysis identified ~2000 proteins (FDR <1%) and, as many as 215 proteins were found to be significantly (p < 0.05) differentially abundant in the 450 vs. 312 mOsm. The effects of high hyperosmolarity (450 mOsm)‐induced damage resembling DES on the HCE cell were further demonstrated by the expression of proteins involved in the activation of inflammatory response (p = 5.5 × 10−3), apoptosis (p = 5.3 × 10−4) and generation of ROS (e.g., glutathione S‐transferase P) (p = 7.1 × 10−3). Moreover, many differentially expressed proteins were also implicated in the regulation of cell viability (p = 1.1 × 10−6) and organization of cytoplasm (e.g., cyclase associated actin cytoskeleton regulatory protein 1) (p = 2.4 × 10−2).Conclusions: In summary, an optimized hyperosmolarity‐induced DES cell culture model for proteomics analysis was established, and specific alterations in the proteome were characterized in this model. This in vitro model would be instrumental for testing therapeutic targets for DES in the future.